Dataset Description

This dataset is designed for training and evaluating Optical Character Recognition (OCR) models for Arabic text. It is an extension of an open-source dataset and includes text rendered in multiple Arabic fonts (Amiri, Sakkal Majalla, Arial, Calibri and Scheherazade New). The dataset simulates real-world book layouts to enhance OCR accuracy.

Dataset Structure

The dataset is divided into five splits based on font name (Sakkal_Majalla, Amiri, Arial, Calibri, and Scheherazade_New). Each split contains data specific to a single font. Within each split, the following attributes are present:

image_name: Unique identifier for each image.
chunk: The text content associated with the image.
font_name: The font used in text rendering.
image_base64: Base64-encoded image representation.

How to Use

from datasets import load_dataset
import base64
from io import BytesIO
from PIL import Image
# Load dataset with streaming enabled
ds = load_dataset("riotu-lab/sample_text2image", streaming=True)
print(ds)




# Load the dataset

# Iterate over a specific font dataset (e.g., Amiri)
for sample in ds["Amiri"]:
    image_name = sample["image_name"]
    chunk = sample["chunk"]  # Arabic text transcription
    font_name = sample["font_name"]
    
    # Decode Base64 image
    image_data = base64.b64decode(sample["image_base64"])
    image = Image.open(BytesIO(image_data))

    # Show the image (optional)
    image.show()

    # Print the details
    print(f"Image Name: {image_name}")
    print(f"Font Name: {font_name}")
    print(f"Text Chunk: {chunk}")
    
    # Break after one sample for testing
    break

OCR Dataset Generation Pipeline

To create your own dataset, you can use the following repository: text2image.